Delirium, consisting of the acute onset of inattention, global cognitive impairment, and other neuropsychiatric symptoms, is a common, serious, and often preventable syndrome among hospitalized and institutionalized elders. At a cost of over $118 billion per year, delirium is associated with adverse outcomes including persistent cognitive decline, loss of independence, and death. A major challenge in advancing treatment trials and prognostic studies has been inadequacies in the current measurement of delirium severity, such as a tendency to overweight hyperactive features or to require highly skilled clinician raters. To address these limitations, our goal is to develop a Delirium Severity Toolkit,a dynamic set of six new measures developed with expert clinical judgment and patient/family/nurse input using modern psychometric theory. Our measures will capture the severity of delirium in various phenomenological presentations (e.g., hypoactive, hyperactive), incorporate multiple perspectives (patient, family member, and nurse), will be useful to a broad array of stakeholders (physicians, nurses, patients, family caregivers, researchers, and policy-makers), and will be developed using state-of-the-art procedures that combine qualitative and quantitative approaches such as those developed and utilized in current major NIH-supported measurement initiatives (e.g., PROMIS, NIH Toolbox, Neuro-QOL). Moreover, our development work will be in compliance with FDA guidance on instrument development to assure applicability for future clinical trials. Our aims are to (1) develop clinical measures of delirium severity that reflect the presence and intensity of multiple dimensions of symptoms; (2) compare these against subjective ratings of delirium severity by patient, family, and nurses; and (3) determine the predictive validity of the measures of delirium severity against adverse clinical outcomes (e.g., functional decline, hospital or post-acute length of stay, institutionalization, healthcare utilization and costs). In the course of instrument development we will demonstrate face, content, and criterion validity, and test responsiveness to change and reliability. To accomplish our aims, we will convene panels of clinical experts, conduct in-depth qualitative interviews with 30 patients, 30 family members, and 30 nurses; develop and refine delirium severity instruments, conduct a field study in 300 patients on 3 services (medicine, surgery, and post-acute care), calibrate and publish final versions (short, long, and computerized adaptive versions) of our 6 delirium severity measures that are ready to use in clinical and research settings. Our approach is innovative since it will fundamentally advance the conceptualization of delirium severity, result in new measures that address limitations in current severity measures, and provide a publicly available item bank for ongoing improvement, validation, and dissemination. We believe that the proposed measurement is essential to advance our understanding of the syndrome of delirium and our ability to effectively treat this important problem.